Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

deepmind.com

Post reply on HN
Print view View original

371–380 of 459 posts

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#371

Post by ygra »

shawnz wrote: post #140
celeritascelery wrote: post #136

Earlier quoted context omitted.

I get what your saying. But screen movement is rate limited (meaning you can't loop through all possible screen positions in 1ms) so you have to actively choose where you want to focus, just like a human player. Think of it more like calls to a web server then "foreach screens".

Can't you click on the minimap to move the camera instantly anywhere on the map? EDIT: I guess you would still have to wait for the next frame to get rendered, which could add up. True, that does change things a bit, but of course a computer could still do that way faster than a human.

They noted that the agent used around 30 viewport changes per minute, about the same as human players.

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#372

Post by gpm »

bitxbitxbitcoin wrote: post #281
gpm wrote: post #215

Earlier quoted context omitted.

> a repetitive strain injury waiting to happen. Yes, I have one from it and wasn't even playing that high (I averaged less than 100 apm). I understand that it's a common problem.

Was Starcraft the only/main game that you played?

Yes, basically the only for several years at that point. A few hours here and their of other games but nothing at all substantial.

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#373

Post by posterboy »

asdfasgasdgasdg wrote: post #189
jgon wrote: post #150

Earlier quoted context omitted.

Even the 200ms reaction time seemed overly slanted towards the AI. I don't think that is the actual reaction time of top pros, in the matches the AI played the human player would teleport in from complete invisibility and try to use an instant cast spell and the AI would have already teleported out. Yes the theoretically may have been constrained to a 200ms reaction time, but in practice the AI was playing at a super…

The game is difficult to watch, but does anyone honestly believe that an AI is going to have a difficult time parsing the scene if it is trained to do so? That to me just seems like a question of resources. We're pretty good at image recognition and segmentation now, and that's without the unlimited amounts of training data one could generate when using a controlled game environment with a limited range of possible a…

You talk about a static image, but navigating the camera requires strategy, attention, and adds to the focus. If you take that away, it's just a turbo charged pen-and-paper RPG with a time limit on rounds.

They could train against the API, reinforcing the AI trying to predict the state from vision. But with limited APM it would be pretty difficult for the AI to keep track of everything. And, potentially, it would still not be the same as a human looking at it. I'm not sure whether human attention is a particularly bad example of efficient resource allocation. I'm very biased to think it is still the gold standard. But the fact that deepmind didn't focus on this implies they were not finding it interesting enough, and/or too difficult.

Anyhow, (visual) exploration is a step up from mere image recognition

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#374

Post by posterboy »

feanaro wrote: post #345
kaveh_h wrote: post #297

Earlier quoted context omitted.

For real life applications, parsing the ”scene” would have impact as it could only convey imperfect information retained. In the game of starcraft the information is perfect when fog of war have been removed this together with unlimited attention (camera viewport) helps action potential and macro planning. No player is ever going to be able to consider precise strategy on the whole map perfectly in their mind. If dee…

There is another potential use: given these limitations, an AI might be able to learn to be better strategically, which could translate to an even greater advantage once the limitations were removed later on.

windowing the focus perhaps, yes, but I'd assume it's the opposite and the focus is applied more freely.

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#375

Post by red75prime »

thomasfortes wrote: post #195
asdfasgasdgasdg wrote: post #189

Earlier quoted context omitted.

The game is difficult to watch, but does anyone honestly believe that an AI is going to have a difficult time parsing the scene if it is trained to do so? That to me just seems like a question of resources. We're pretty good at image recognition and segmentation now, and that's without the unlimited amounts of training data one could generate when using a controlled game environment with a limited range of possible a…

But on the other hand, an AI that beats humans using brute force in a game where it makes a ton of difference isn't much fair too.

> using brute force

"Brute force" in AI context is usually reserved for traversal of the entire search space. I think "superhuman micromanagement" is a better term. And before AlphaStar superhuman micro wasn't insurmountable obstacle for human players.

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#376

Post by ece »

I think Deepmind definitively showed the agents are learning high-level play, which was great to see. I didn't really pay attention to AlphaGo, but did skim the AlphaZero paper, and I'm not really left with any doubts about how good RL/LSTMs can get against other AI or humans given enough time to train.

That said, it's an open question whether given all the constraints of the last live match and what was mentioned during the talk (that newer strategies keep getting discovered by humans and agents), whether humans could even win 50% of the time against an agent.

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#377

Post by jon-wood »

sleepydog wrote: post #63
modeless wrote: post #8

AlphaStar interacted with the StarCraft game engine directly via its raw interface, meaning that it could observe the attributes of its own and its opponent’s visible units on the map directly, without having to move the camera - effectively playing with a zoomed out view of the game Additionally, and subsequent to the matches, we developed a second version of AlphaStar. Like human players, this version of AlphaStar…

I'm curious, would the AI be able to see cloaked units? In sc1 you could see them,( I think sc2 is the same) but it was very difficult. How does the 'raw' interface expose that subtlety?

My understanding is that the AI sees things via an API the game exposes, so presumably cloaked units are completely invisible to it until they're revealed.

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#378

Post by jon-wood »

minimaxir wrote: post #20
b_tterc_p wrote: post #7

Earlier quoted context omitted.

https://seaborn.pydata.org/examples/kde_ridgeplot.html I would try with a Seaborn facet grid? I think they’ve got something custom, but this should get close (be aware this specific example is a kde so it will normalize total area) On a second glance this won’t get the 3D horizontal offset.

A lot of the viz styles seem close to ggplot2 which is why I though it was made using ggridges ( https://cran.r-project.org/web/packages/ggridges/vignettes/i... ) but I'm not sure if you can do a perceptual skew in ggplot2.

It wouldn't surprise me if they generated graphs and then handed them off to a design team to trace and make pretty.

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#379

Post by stared »

rcheu wrote: post #46

This is really impressive, I didn't expect starcraft to be played this well by a machine learning based AI. I'm excited to read the paper when it comes out! That said, I'm not sure I agree that it was winning mainly due to better decision making. For context, I've been ranked in the top 0.1% of players and beaten pros in Starcraft 2, and also work as a machine learning engineer. The stalker micro in particular looked…

APM is one thing. I am curious what would happen if it could only see a limited view (as in the last game with MaNa, which it lost to him) and physical click dynamics (i.e. clicking + gaussian noise as an action, instead of giving direct commands). That way there will be misclicks, preventing this super-efficient Stalker micro.

Re: AlphaStar: Mastering the Real-Time Strategy Game StarCraft II

#380

Post by porky »

rcheu wrote: post #46

This is really impressive, I didn't expect starcraft to be played this well by a machine learning based AI. I'm excited to read the paper when it comes out! That said, I'm not sure I agree that it was winning mainly due to better decision making. For context, I've been ranked in the top 0.1% of players and beaten pros in Starcraft 2, and also work as a machine learning engineer. The stalker micro in particular looked…

And also, latency is lower
Post reply on HN